PERSPECTIVES FOR FUTURE LARGE EARTHQUAKES IN THE SISZ

As discussed earlier the moment released in the two large earthquakes of 2000 is estimated to be $1.2\ast10^{19}$Nm, while the the moment built up and released during a 140 year earthquake cycle has been estimated to be 0.7- $1\ast10^{20}$Nm, where the higher value is based on the estimated size of historical earthquakes. Assuming that the lower value is more realistic, as the historical earthquake magnitudes may have been overestimated, and taking into account that only 100 years have apparently elapsed of the 140 year cycle (Stefánsson and Halldórsson 1988), the moment build-up before the earthquakes would have been $5\ast10^{19}$ Nm. This means that only a fourth of the stored moment would have been released in the two large earthquakes in 2000. The remaining moment is probably mostly stored in the easternmost part of the SISZ, where the largest earthquakes are to be expected as the elastic/brittle crust is thickest there. Judging from historical observations and the general understanding of tectonics outlined above, the build-up of strain from around 1900 to year 2000 has not been enough to produce a magnitude 7 earthquake in the easternmost part of the zone. We suggest, however, that further build-up of strain, in addition to what remains after the recent earthquakes will be enough to rupture the strong crust there within the next few decades. The above reasoning is based on a simple model of moment build-up, assuming steady plate motions with shearing deformation across a homogeneous SISZ, however, with increasing thickness and strength from west to east. The total release of stress in such a simplified zone would have the tendency to delay until it starts at the easternmost, strongest part and trigger subsequent earthquakes further west during a relatively short time frame.

Although there is some historical support for this hypothesis, both history and the recent events show deviations from such a simple model. The deviations may result from stress heteorogeneities within the zone causing fracture criticality to be reached locally, before stresses reach the fracture criticality of the zone as a whole.

It has also been proposed that strain build-up for earthquakes in this area is not only due to general plate motion, but also has a local build-up of stress, possibly caused by intrusion of fluids near the bottom of the seismogenic crust (Stefánsson and Halldórsson 1988) Considering this, it is suggested that it is still possible in the present cycle that an earthquake of a comparable size to the recent earthquakes may also occur farther west, either before a possible magnitude 7 earthquake in the easternmost part, or triggered by stress transfer by such an earthquake.